|Veröffentlichungsdatum||24. März 1987|
|Eingetragen||29. Sept. 1983|
|Prioritätsdatum||30. Sept. 1982|
|Veröffentlichungsnummer||06536875, 536875, US 4652940 A, US 4652940A, US-A-4652940, US4652940 A, US4652940A|
|Ursprünglich Bevollmächtigter||Hitachi, Ltd.|
|Zitat exportieren||BiBTeX, EndNote, RefMan|
|Patentzitate (13), Referenziert von (41), Klassifizierungen (13), Juristische Ereignisse (4)|
|Externe Links: USPTO, USPTO-Zuordnung, Espacenet|
This invention relates to on-line data processing systems and more particularly to the control for reading a magnetic tape for a recorded journal in which a hysteresis of process is recorded, at the time of a system resuming its processing after the system has failed and was restored to the normal state.
In an on-line data processing system, it is a general practice that the historical information (journal) of the processes executed therein is recorded by a magnetic tape unit.
The journal thus obtained is used chiefly for the system resuming process after the system has failed and was restored to the normal state.
The on-line data processing system performs such processes as reading a file (collected information for an object stored in a magnetic disk unit, magnetic tape unit or the like), updating the file and so on. If a failure occurs in the hardware or software of the system, the process may be stopped. At this time, the failure is removed by proper processing and the system is again started to operate. Before execution of a process, a process (system resuming process) is necessary to make the contents of the file appropriate by using the journal. In addition, to read the journal, it is absolutely necessary to fully check the magnetic tape loaded on the magnetic tape unit for a desired journal recorded therein, because the operator may have loaded an improper magnetic tape on the unit.
In general, checking of what information is recorded on the magnetic tape has so far been performed by reading out information, calling label, recorded at the beginning of the tape.
Furthermore, to restore the file to a proper state, it is necessary to first read the most recent journal.
Therefore, after the magnetic tape is rewound to its beginning where the label is read, and the presence of a proper journal is confirmed, the tape must be wound forward until the most recent journal recorded portion is read out. For this process, a time of the order of minutes is required. This means that the start of process execution is delayed by that time.
Accordingly, it is an object of the invention to quickly determine that the magnetic tape being read has a proper journal recorded therein.
According to this invention, information is provided indicative of whether the magnetic tape has a proper journal recorded therein, as part of each journal data record, and in the system resuming process, checking is made of the journal closest to the head position at the time of system failure.
FIG. 1 is a schematic diagram of an on-line system;
FIG. 2 is a flow chart of a process forming one embodiment of this invention;
FIG. 3 shows an example of a journal in one embodiment of this invention;
FIG. 4 is a flow chart of the process for journal reading preparation in FIG. 2;
FIG. 5a shows one example of the state of a main memory unit within a central processing unit in one embodiment of this invention;
FIG. 5b shows one example of a file in one embodiment of this invention;
FIG. 6 shows a general format of the label on the magnetic tape;
FIG. 7 shows one example of a journal data record in one embodiment of this invention;
FIGS. 8, 9 and 10 show other examples of a journal data record in one embodiment of this invention; and
FIGS. 11a to 11d show examples of the state in which journals have been obtained.
One embodiment of the invention in such an on-line system as is used in banks will be described below.
FIG. 1 is a schematic block diagram of an on-line system. Referring to FIG. 1, there are shown terminals 1 corresponding to the window equipment or automatic money-receiving and -repaying equipment installed at each branch in the on-line system of banks. These terminals 1 are connected to a central processing unit 2 through communications channels. The central processing unit 2 is also connected to magnetic disk equipment 3 in which data (file) for each customer is stored, and to magnetic tape equipment 4 in which a journal is stored. FIG. 2 shows a process flow to be executed in the on-line system. Before describing this process, we will explain examples of a journal and a file. FIG. 3 shows an example of a magnetic tape for a journal according to this invention. As in the general magnetic tape for a journal, label information 21 is stored on the beginning portion of a magnetic tape 20 and many journal data records 22 are stored in turn in the following tape portion. The format of the journal data record 22 according to this invention, as illustrated, has control information from block length to date and the following journal data. However, there is sometimes no journal as will be described later. One of the features of this invention is to have ID23 indicative of a journal data record and the types of journal and date 24 indicative of the data of the journal acquisition.
FIG. 5a shows the state in which information is stored in a main memory 50 within the central processing unit 2. Numeral 51 represents a program for executing the on-line process of FIG. 2. The instructions of this program are sequentially read and executed to operate the central processing unit 2, in performing the process. The program for the system resuming process is stored in the magnetic disk unit. When the system resuming process is required to be executed, the program is loaded in the main memory unit 50. The data inputted from the terminals 1, data read from the file, data after processing and so on are stored in the main memory unit 50.
FIG. 5b shows the deposit file stored in the magnetic disk unit 3, which includes account number 57 and balance 58.
FIG. 6 illustrates the general format of a label including the volume serial number, the data set name, and the above-mentioned ID and date.
Now, the on-line data process will be described with reference to FIG. 2. Discrimination indicator (-) as a request for withdrawal of funds, account number for the withdrawal and the amount of money to be paid are transferred as input data 52 from terminal 1 to central processing unit 2 and stored in the main memory unit 50.
The input data 52 is delivered to the in-line data process program 51 on the central processing unit 2 together with the window or terminal name. In the on-line data process, the account number is obtained from the input data 52 (at step 101). Then, a deposit file 3 is referred to and the corresponding account data 53 is read (at step 102).
In order to update the account data 53, the data 53 is transferred to area 54 and the amount to be paid for the input data 52 is subtracted therefrom (at step 103). If the balance of the updated information 54 is less than zero under which the withdrawal of deposit is impossible, an error response to the window equipment 1 is generated (step 112) and response data is transmitted (step 113).
If the balance of the updated information 54 is larger than zero, under which condition the withdrawal is possible, the updated information 54 is written in the deposit file in place of the nonupdated information 53 and the updated information 54 is edited in the format shown in FIG. 7 and stored as journal data 56 in the magnetic tape unit 4 (step 105). This is called FJ (file journal).
The file journal of FIG. 7, of course, follows the format of the journal data record and includes the deposit file name, account number, non-processed information 26 (nonupdated information 53), and processed information 27 (updated information 54).
Thereafter, the updated information 54 is written in the file 3 to thereby update the file 3 (step 106).
After completion of updating the deposit file 3, normal response 55 to the terminal 1 is edited (step 107). This normal response 55 is also edited together with the terminal name in the format shown in FIG. 9 and written in the journal 4 (step 108). This is called OJ (output journal).
Then, the journal indicative of the completion of one on-line data process is edited in the format shown in FIG. 8 and written in the journal 4. This is called PJ (process journal).
Finally, the normal response 55 is transmitted to the terminal 1. When the transmission is completed, the journal to the effect that the response data transmission is completed is edited in the format shown in FIG. 10 and is obtained. This is called AJ (acknowledgement journal). These 4 different journals are stored in the order of their occurrence and discriminated by the ID as described above.
The terminal then pays money in accordance with the response.
The condition of the on-line system at the time of failure will be described with reference to FIGS. 2, 4 and 11a to 11d.
When the system is resumed after occurrence of system failure, it is a general practice that the operator directs the system to resume process through the system control. If this is done, a system resume process program is loaded into the main memory unit, and executed (at step 115). First, the preparation for reading the journal is made. This process is important in this invention, as illustrated in detail in FIG. 4. First, backward reading is performed from the record closest to the magnetic head (at step 30). That is, data is read from the magnetic tape running backward, or in the opposite direction to the writing direction, as is already known; or a sequence of information written is read from the back portion to the front portion.
Then, it is checked whether the read data is present or not (step 31). If the data is present, checking is made of whether the magnetic tape is journal data set, by the ID information 23 within the data (step 32). If the decision is YES, it is checked by the data information 24 whether the data is data which is to be processed and which was obtained on the appointed day (step 33). Since it is confirmed at the end of each day that the completion of process and file are correct, the journal data before the previous day is not necessary.
If the result of the checking is YES, the checking of the magnetic tape for propriety is finished. If the decision either at step 32 or 33 is NO, the journal record therebefore is read in by backward reading, and the magnetic tape is checked for propriety. If at step 31 no data is detected, the central processing unit supplies an error message to the display unit or the like (step 34). The operator, when looking at the display, sets another magnetic tape in the tape unit.
The magnetic tape is then checked for propriety again in accordance with the above procedure.
It will be understood from the above description that it is not necessary to rewind to the label and then wind to the newest journal record so that the process can be performed at high speed.
Then, the journal is read in (step 116), and the resuming process proceeds.
The on-line system may go down at any step in the on-line data process. If it occurs between steps 106 and 110 in FIG. 2, there is caused the situation that the balance in the deposit file 3 is updated but no money is paid by the terminal.
In order to avoid such a condition, the system resuming process determines when the on-line system has gone down, in accordance with the contents of the journal 4 and removes it. The method for determination will be described below.
At the time of the process for resuming the on-line system operation, the journal takes one of four possible states which are identified at decision steps 118 to 121. The state of FIG. 11a occurs after the step 111 or before the step 106 and no contradiction is caused, therefore nothing is done. The state of FIG. 11b occurs between the steps 109 and 111 of FIG. 2 and there the system has gone down at a time when the response data 55 is not transmitted to the terminal 1. Thus, to prepare for the request of resending the response data by the operator of the terminal, the response data on OJ 85 is read in as response data 55.
The state of FIG. 11c occurs between the steps 108 and 109 of FIG. 2 and the newest record ID shows OJ. Since PJ is not obtained, the on-line data process is treated as being incomplete. In other words, since the update of the deposit file 3 at step 106 is invalid, the state before update is returned to on FJ 87 at step 124 of FIG. 2.
The condition of FIG. 11d is caused between the steps 106 and 108 in FIG. 2 and shows a case where the system has gone down. First, FJ is detected. Since PJ is not detected, the on-line data process is treated as being incomplete. That is, since the update of the deposit file at step 106 is invalid, the state before update is returned to on FJ 89 at step 124 in FIG. 2.
The process for resuming the system operation is finished.
Thereafter, the on-line data process program 51 is again executed.
|US3321747 *||2. Okt. 1964||23. Mai 1967||Hughes Aircraft Co||Memory protection system|
|US3801963 *||28. Sept. 1972||2. Apr. 1974||Burroughs Corp||Method and apparatus for transferring data from a volatile data store upon the occurrence of a power failure in a computer|
|US3852571 *||16. Aug. 1972||3. Dez. 1974||Hempstead Bank||System of transferral of funds|
|US3904860 *||26. Nov. 1973||9. Sept. 1975||Siemens Ag||Method for diagnosing the condition of data processors|
|US3959778 *||30. Aug. 1974||25. Mai 1976||Compagnie Honeywell Bull (Societe Anonyme)||Apparatus for transferring data from a volatile main memory to a store unit upon the occurrence of an electrical supply failure in a data processing system|
|US3973237 *||14. Juni 1974||3. Aug. 1976||Fujitsu Ltd.||Article depositing machine|
|US4306299 *||13. Nov. 1979||15. Dez. 1981||Pitney-Bowes, Inc.||Postage meter having means transferring data from a working memory to a non-volatile memory under low power conditions|
|US4316248 *||5. Nov. 1979||16. Febr. 1982||Data General Corporation||Memory refresh means including means for providing refresh addresses during power failures|
|US4355369 *||15. Juni 1979||19. Okt. 1982||Docutel Corporation||Automatic banking machine|
|US4404649 *||3. Nov. 1980||13. Sept. 1983||Recognition Equipment Incorporated||Document processing system|
|US4458307 *||17. Jan. 1980||3. Juli 1984||Burroughs Corporation||Data processor system including data-save controller for protection against loss of volatile memory information during power failure|
|US4517663 *||22. Dez. 1983||14. Mai 1985||Fujitsu Fanuc Limited||Method of rewriting data in non-volatile memory, and system therefor|
|US4523295 *||7. Sept. 1982||11. Juni 1985||Zenith Electronics Corporation||Power loss compensation for programmable memory control system|
|Zitiert von Patent||Eingetragen||Veröffentlichungsdatum||Antragsteller||Titel|
|US5062046 *||20. März 1990||29. Okt. 1991||Hitachi, Ltd.||Multiple processor system having a correspondence table for transferring processing control between instruction processors|
|US5089958 *||23. Jan. 1989||18. Febr. 1992||Vortex Systems, Inc.||Fault tolerant computer backup system|
|US5372507 *||11. Febr. 1993||13. Dez. 1994||Goleh; F. Alexander||Machine-aided tutorial method|
|US5649152||13. Okt. 1994||15. Juli 1997||Vinca Corporation||Method and system for providing a static snapshot of data stored on a mass storage system|
|US5835953||8. Nov. 1996||10. Nov. 1998||Vinca Corporation||Backup system that takes a snapshot of the locations in a mass storage device that has been identified for updating prior to updating|
|US5901004 *||22. Sept. 1997||4. Mai 1999||Sony Corporation||Recording and reproducing apparatus, information signal recording and reproducing system and method of managing invalid area information|
|US6181498||20. Jan. 1995||30. Jan. 2001||Sony Corporation||Recording and reproducing apparatus, information signal recording and reproducing system and method of managing invalid area information|
|US6301588 *||6. Febr. 1996||9. Okt. 2001||Sony Corporation||Signal processing method|
|US7239581||24. Aug. 2004||3. Juli 2007||Symantec Operating Corporation||Systems and methods for synchronizing the internal clocks of a plurality of processor modules|
|US7272666||13. Febr. 2004||18. Sept. 2007||Symantec Operating Corporation||Storage management device|
|US7287133||24. Aug. 2004||23. Okt. 2007||Symantec Operating Corporation||Systems and methods for providing a modification history for a location within a data store|
|US7296008||24. Aug. 2004||13. Nov. 2007||Symantec Operating Corporation||Generation and use of a time map for accessing a prior image of a storage device|
|US7409587||24. Aug. 2004||5. Aug. 2008||Symantec Operating Corporation||Recovering from storage transaction failures using checkpoints|
|US7536583||13. Okt. 2006||19. Mai 2009||Symantec Operating Corporation||Technique for timeline compression in a data store|
|US7577806||23. Sept. 2003||18. Aug. 2009||Symantec Operating Corporation||Systems and methods for time dependent data storage and recovery|
|US7577807||24. Aug. 2004||18. Aug. 2009||Symantec Operating Corporation||Methods and devices for restoring a portion of a data store|
|US7584337||13. Febr. 2004||1. Sept. 2009||Symantec Operating Corporation||Method and system for obtaining data stored in a data store|
|US7631120||24. Aug. 2004||8. Dez. 2009||Symantec Operating Corporation||Methods and apparatus for optimally selecting a storage buffer for the storage of data|
|US7725667||12. März 2004||25. Mai 2010||Symantec Operating Corporation||Method for identifying the time at which data was written to a data store|
|US7725760||24. Aug. 2004||25. Mai 2010||Symantec Operating Corporation||Data storage system|
|US7730222||24. Aug. 2004||1. Juni 2010||Symantec Operating System||Processing storage-related I/O requests using binary tree data structures|
|US7827362||24. Aug. 2004||2. Nov. 2010||Symantec Corporation||Systems, apparatus, and methods for processing I/O requests|
|US7904428||24. Aug. 2004||8. März 2011||Symantec Corporation||Methods and apparatus for recording write requests directed to a data store|
|US7991748||17. Febr. 2004||2. Aug. 2011||Symantec Corporation||Virtual data store creation and use|
|US8521973||28. Sept. 2007||27. Aug. 2013||Symantec Operating Corporation||Systems and methods for providing a modification history for a location within a data store|
|US20050063374 *||12. März 2004||24. März 2005||Revivio, Inc.||Method for identifying the time at which data was written to a data store|
|US20050065962 *||17. Febr. 2004||24. März 2005||Revivio, Inc.||Virtual data store creation and use|
|US20050066118 *||24. Aug. 2004||24. März 2005||Robert Perry||Methods and apparatus for recording write requests directed to a data store|
|US20050066222 *||23. Sept. 2003||24. März 2005||Revivio, Inc.||Systems and methods for time dependent data storage and recovery|
|US20050066225 *||24. Aug. 2004||24. März 2005||Michael Rowan||Data storage system|
|US20050076261 *||13. Febr. 2004||7. Apr. 2005||Revivio, Inc.||Method and system for obtaining data stored in a data store|
|US20050076264 *||24. Aug. 2004||7. Apr. 2005||Michael Rowan||Methods and devices for restoring a portion of a data store|
|US20060047895 *||24. Aug. 2004||2. März 2006||Michael Rowan||Systems and methods for providing a modification history for a location within a data store|
|US20060047902 *||24. Aug. 2004||2. März 2006||Ron Passerini||Processing storage-related I/O requests using binary tree data structures|
|US20060047903 *||24. Aug. 2004||2. März 2006||Ron Passerini||Systems, apparatus, and methods for processing I/O requests|
|US20060047925 *||24. Aug. 2004||2. März 2006||Robert Perry||Recovering from storage transaction failures using checkpoints|
|US20060047989 *||24. Aug. 2004||2. März 2006||Diane Delgado||Systems and methods for synchronizing the internal clocks of a plurality of processor modules|
|US20060047998 *||24. Aug. 2004||2. März 2006||Jeff Darcy||Methods and apparatus for optimally selecting a storage buffer for the storage of data|
|US20060047999 *||24. Aug. 2004||2. März 2006||Ron Passerini||Generation and use of a time map for accessing a prior image of a storage device|
|US20070088973 *||13. Okt. 2006||19. Apr. 2007||Revivio, Inc.||Technique for timeline compression in a data store|
|US20090019459 *||28. Sept. 2007||15. Jan. 2009||Symantec Operating Corporation||Systems and methods for providing a modification history for a location within a data store|
|US-Klassifikation||360/5, 714/E11.13, 235/379, 705/30|
|Internationale Klassifikation||G06F7/22, G06F12/00, G06F15/00, G06F3/06, G06F11/14|
|Europäische Klassifikation||G06F11/14A12, G06Q40/10|
|29. Sept. 1983||AS||Assignment|
Owner name: HITACHI, LTD., 6, KANDA SURUGADAI 4-CHOME, CHIYODA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:SUMIYOSHI, TAKASHI;REEL/FRAME:004180/0324
Effective date: 19830914
|2. Juli 1990||FPAY||Fee payment|
Year of fee payment: 4
|1. Juli 1994||FPAY||Fee payment|
Year of fee payment: 8
|31. Aug. 1998||FPAY||Fee payment|
Year of fee payment: 12